Device and method using a neural network to detect and compensate an air vacuum effect

Abstract

Device and method using a neural network to detect and compensate an air vacuum effect. The device stores a predictive model comprising weights of a neural network. The device receives an area temperature measurement (representative of a temperature of an area where the device is located) from a temperature sensing module of the device. The device determines at least one other measurement related to the device. The device executes a neural network inference engine implementing a neural network, using the predictive model for inferring output(s) based on inputs. The inputs comprise the area temperature measurement and the at least one other measurement related to the device. The output(s) comprises a metric representative of an air vacuum effect in the device. The device determines if an adjustment of the area temperature measurement needs to be performed based on the metric representative of the air vacuum effect in the device.

What is claimed is: 
     
         1 . A device comprising:
 memory for storing a predictive model comprising weights of a neural network;   a temperature sensing module adapted for measuring a temperature of an area where the device is located; and   a processing unit comprising one or more processor configured to:
 receive an area temperature measurement from the temperature sensing module; 
 determine at least one other measurement related to the device; and 
 execute a neural network inference engine, the neural network inference engine implementing a neural network using the predictive model for inferring one or more output based on inputs, the inputs comprising the area temperature measurement and the at least one other measurement related to the device, the one or more output comprising a metric representative of an air vacuum effect in the device. 
   
     
     
         2 . The device of  claim 1 , wherein the processing unit further determines if an adjustment of the area temperature measurement needs to be performed based on the metric representative of the air vacuum effect in the device. 
     
     
         3 . The device of  claim 1 , wherein the processing unit includes an integrated temperature sensing module and the at least one other measurement related to the device comprises a temperature measurement of the processing unit performed by the integrated temperature sensing module. 
     
     
         4 . The device of  claim 1 , further comprising at least one other temperature sensing module, each other temperature sensing module performing an internal temperature measurement; and wherein determining at least one other measurement related to the device comprises receiving each internal temperature measurement from the corresponding other temperature sensing module. 
     
     
         5 . The device of  claim 4 , wherein each internal temperature measurement is representative of heat generated by at least one of the following components of the device: a display of the device, a communication interface of the device, the processing unit, a power unit of the device and an electrical load controller of the device. 
     
     
         6 . The device of  claim 1 , wherein the at least one other measurement related to the device comprises a utilization metric of the one or more processor of the processing unit. The device of  claim 6 , wherein the utilization metric of the one or more processor of the processing unit is calculated based on a Central Processing Unit (CPU) utilization of each of the one or more processor. 
     
     
         8 . The device of  claim 1 , wherein the at least one other measurement related to the device comprises a utilization metric of a display of the device. 
     
     
         9 . The device of  claim 8 , wherein the utilization metric of the display is representative of a dimming level of the display, a light intensity output of the display, a backlight level of the display or an illumination in the area where the device is located, the illumination in the area being measured by an illumination sensor of the device. 
     
     
         10 . The device of  claim 1 , wherein the at least one other measurement related to the device comprises at least one of the following: a utilization metric of a communication interface of the device and a utilization metric of a power unit of the device. 
     
     
         11 . The device of  claim 1 , wherein the neural network inference engine implements a neural network comprising an input layer, followed by fully connected hidden layers, followed by an output layer; the input layer comprising neurons respectively receiving the area temperature measurement and the at least one other measurement related to the device;
 the output layer comprising a neuron outputting the metric representative of an air vacuum effect in the device; the weights of the predictive model being applied to the fully connected hidden layers.   
     
     
         12 . The device of  claim 1 , consisting of a smart thermostat. 
     
     
         13 . A method using a neural network to detect and compensate an air vacuum effect, the method comprising:
 storing a predictive model comprising weights of a neural network in a memory of a device;   receiving by a processing unit of the device an area temperature measurement from a temperature sensing module of the device, the area temperature measurement being representative of a temperature of an area where the device is located;   determining by the processing unit at least one other measurement related to the device; and   executing by the processing unit a neural network inference engine, the neural network inference engine implementing a neural network using the predictive model for inferring one or more output based on inputs, the inputs comprising the area temperature measurement and the at least one other measurement related to the device, the one or more output comprising a metric representative of an air vacuum effect in the device.   
     
     
         14 . The method of  claim 13 , further comprising determining by the processing unit if an adjustment of the area temperature measurement needs to be performed based on the metric representative of the air vacuum effect in the device. 
     
     
         15 . The method of  claim 13 , wherein determining by the processing unit at least one other measurement related to the device comprises determining a temperature measurement of the processing unit performed by an integrated temperature sensing module included in the processing unit. 
     
     
         16 . The method of  claim 13 , wherein determining by the processing unit at least one other measurement related to the device comprises receiving at least one internal temperature measurement, each internal temperature measurement being respectively performed by another temperature sensing module included in the device. 
     
     
         17 . The method of  claim 16 , wherein each internal temperature measurement is representative of heat generated by at least one of the following components of the device: a display of the device, a communication interface of the device, the processing unit, a power unit of the device and an electrical load controller of the device. 
     
     
         18 . The method of  claim 13 , wherein determining by the processing unit at least one other measurement related to the device comprises determining a utilization metric of the one or more processor of the processing unit. 
     
     
         19 . The method of  claim 18 , wherein the utilization metric of the one or more processor of the processing unit is calculated based on a Central Processing Unit (CPU) utilization of each of the one or more processor. 
     
     
         20 . The method of  claim 13 , wherein determining by the processing unit at least one other measurement related to the device comprises determining a utilization metric of a display of the device. 
     
     
         21 . The method of  claim 20 , wherein the utilization metric of the display is representative of a dimming level of the display, a light intensity output of the display, a backlight level of the display or an illumination in the area where the device is located, the illumination in the area being measured by an illumination sensor of the device 
     
     
         22 . The method of  claim 13 , wherein determining by the processing unit at least one other measurement related to the device comprises determining at least one of the following: a utilization metric of a communication interface of the device and a utilization metric of a power unit of the device. 
     
     
         23 . The method of  claim 13 , wherein the neural network inference engine implements a neural network comprising an input layer, followed by fully connected hidden layers, followed by an output layer; the input layer comprising neurons respectively receiving the area temperature measurement and the at least one other measurement related to the device;
 the output layer comprising a neuron outputting the metric representative of an air vacuum effect in the device; the weights of the predictive model being applied to the fully connected hidden layers.   
     
     
         24 . The method of  claim 13 , wherein the device consists of a smart thermostat. 
     
     
         25 . A non-transitory computer program product comprising instructions executable by a processing unit of a computing device, the execution of the instructions by the processing unit providing for using a neural network to detect and compensate an air vacuum effect by:
 storing a predictive model comprising weights of a neural network in a memory of a device;   receiving by a processing unit of the device an area temperature measurement from a temperature sensing module of the device, the area temperature measurement being representative of a temperature of an area where the device is located;   determining by the processing unit at least one other measurement related to the device; and   executing by the processing unit a neural network inference engine, the neural network inference engine implementing a neural network using the predictive model for inferring one or more output based on inputs, the inputs comprising the area temperature measurement and the at least one other measurement related to the device, the one or more output comprising a metric representative of an air vacuum effect in the device.   
     
     
         26 . The computer program product of  claim 25 , the execution of the instructions further provide for determining by the processing unit if an adjustment of the area temperature measurement needs to be performed based on the metric representative of the air vacuum effect in the device.